OpenStax-CNX module: m50415
1
What is Logical or Rational
Thinking, and how does it relate to
Reasoning, Heuristics, Biases and
∗
the Rationality Debate?
Mark Pettinelli
This work is produced by OpenStax-CNX and licensed under the
†
Creative Commons Attribution License 4.0
There are two dierent types of intelligence - one type can be measured objectively (i.e. perceptual speed
and memory), and the type of intellect is subjective and, although it can be measured, is still subjective.
The subjective type of intellect consists of things like reasoning ability and verbal comprehension.
I stated that something like verbal comprehension is subjective; however that statement is actually a big
idea (if you think about it). It is basically saying that every words denition is up for debate, or subject
to opinion. That is true, however - for instance the meaning of each word for each person may be dierent.
When someone says the word 'dog' maybe they mean to use the word as a metaphor and really mean, 'that
person is like a dog' not 'that is a dog'. Maybe even when someone says 'that is a dog' they are making
a subjective statement, even though it seems pretty objective. - I mean a dog has a strict denition and
most people have the same thing in mind when they think of that word, therefore making its meaning rather
straightforward.
My point is that dierent kinds of emotional understanding (which are largely things in life that are
'subjective') make up life, the words people use, and common human understanding. Therefore nothing is
ever really 'objective' because it is subject to human biases. Mathematical equations are objective, however
if a animal were to look at a math problem they might not understand it as being objective - they might
interpret the problem to mean something else (since it wouldn't mean anything to them mathematically).
People have beliefs of various sorts. These beliefs inuence their thinking and how they feel.
What else is to be said about subjective reasoning?
What else is to be said about what I have called 'subjective reasoning' ? I am labeling reasoning ability as
being biased and subjective in any case where emotional information is handled, which is all the time unless
something is completely objective. However, nothing is completely objective because even a math problem
is going to cause someone to be emotional or process it emotionally in some way. That is why I am saying
that all reasoning ability is actually a sort of 'subjective reasoning'.
I mean, if you think about it, most if not all of life involves dealing with your own personal feelings whether you are aware of it or not. Feelings are always present, they bias your decisions, and they motivate
your behaviors and thoughts.
∗
†
Version 1.17: Jul 24, 2014 2:47 am -0500
http://creativecommons.org/licenses/by/4.0/
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
2
Feelings eect our lives
How is someone supposed to know when their feelings or other ideas they have (such as a belief about
something) inuence their decisions or thinking?
Is most of thinking emotional and biased? Or is most of the thinking people do fairly straightforward
and not involve making complex (and potentially inuenced by feeling) decisions?
the most emotionally relevant factor is the motivator
Goals can be changed by how motivated someone is to have that goal.
1
conscious awareness at various times for various reasons. Simon (1967)
Some goals can be brought into
reasoned that emotions are calls
for reprioritization: that emotion regarding a goal that is out of awareness eventually induces people to give
that goal a higher priority. The stronger the emotion, the stronger the claim for higher priority. Aect pulls
the out-of-awareness into awareness.
Simons analysis was just referring to goals.
However, if you think about it, all of someones thoughts
might follow a similar logic - the logic being that the most emotionally relevant thought has the highest
claim to priority.
So if someone wants something, then they are emotionally motivated to think certain things because
thinking those things will generate more pleasurable emotions.
computational components underlying intelligence
What are the computational components underlying intelligence?
To begin, I ask the question - is thinking straightforward or is it complicated?
When people think, they are constantly making emotional assessments of various sorts. They think about
their own motivations, i.e. how they feel about dierent things, and what their goals are going to be based
on those motivations.
Individual thoughts also mean something emotionally. Anything someone thinks is going to be associated
with dierent feelings and preferences.
Does this mean that thinking is simple and logical?
People think all of the time, what guides their
thoughts are emotional preferences that were formed from previous development or at birth.
1 Semantics versus Cognitive Representations
Louis Narens
2
presents the idea that there is a dierence between descriptive semantics (the words people
use to describe something) and cognitive representations (which is basically the image or idea your mind
makes up in your head (kind of like an abstract thought)) in evaluating evidence for judgments:
Support Theory has an empirical base of results showing that dierent descriptions of the same event
often produce dierent subjective probability estimates. It explains these results in terms of subjective
evaluations of supporting evidence. It assumes that events are evaluated in terms of subjective evidence
invoked by their descriptions, and that the observed numerical probability judgments are the result of
the combining of such evaluations of support in a manner that is consistent with a particular equation.
The processes of evaluation are assumed to employ heuristics like those of Kahneman and Tversky, and
because of this, are subject to the kinds of biases introduced by such heuristics.
This article provides a New Foundation for Support Theory. The New Foundation makes a sharp distinction between semantical representations of descriptions as part of natural language processing and
cognitive representations of descriptions as part of a probabilistic judgment. In particular, judgments
of probability employ a complementation operation that has no counterpart in the semantics.
The
complementation operation is used to construct cognitive events that are employed in the computation
of the estimated probability.
So when someone evaluates a piece of information, they describe it in their mind (unconsciously or unconsciously) with words. Then they probably come to a conclusion from the evidence that the description
provided.
1 Simon,
2 A New
H. A. (1967). Motivational and emotional controls of cognition. Psychology Review, 74, 29-39.
Foundation for Support Theory. (2004) Louis Narens. University of California, Irvine
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
3
So describing something with words would be something like, "Linda is a bank teller", or "Linda is a
bank teller and is active in the feminist movement" Here is the explanation from Narens:
Kahneman and Tversky found that over 85% of participants believed it was more likely that Linda
was both a bank teller and a feminist than just a bank teller. This is an example of what has become
known as the conjunction fallacy. According to Kahneman and Tversky, it is due to representativeness:
bank teller and is active in the feminist movement is more a representative description of Linda
than just bank teller.
So a humans mind has the verbal description given to them in words, and then their mind forms a representation based o of what they heard (i.e. - possibly an idea of Linda in their minds).
So that means that there must be lots of words use people use to describe things, and also lots of cognitive
'ideas' or 'representations' they have in their mind that might assist these words.
So words, ideas and representations are all things a human's mind uses to think.
I don't know when
exactly a human mind might use words instead of abstract, non-verbal thoughts - that would be getting
unnecessarily detailed into how thinking works, I would say.
2 So what exactly is a 'Sub jective Evaluation' ?
A subjective evaluation is exactly what those words describe - an assessment or evaluation of something that
is biased, opinionated, and even possibly highly inuenced by the persons feelings.
Subjective evaluations are important because people make them all of the time, whether or not they
are aware of it. For instance anytime you see another person your mind makes an opinionated assessment
of them. You might or might not be aware of your unconscious assessment - maybe you make a conscious
assessment of the person that is dierent from your unconscious one, in which case you could feel confused
about the person or something.
Since earlier in this paper I stated that everything in life is actually subjective, that means that people
are constantly making subjective evaluations whenever they think about anything.
Any thought about
something could be subjective in some way.
If you see a photograph maybe you have an unconscious opinion of that - or even if you think about
something you wouldn't typically consider to be emotional your mind could still have a strong unconscious
feeling or interpretation.
3 How to develop a logical reasoner
The human mind (and animal minds, though the process is dierent) comes to conclusions by weighing
evidence. This process could be done unconsciously or consciously; for instance people might make if - then
statements to think about material. Part of that might be considering evidence from examples that easily
come to mind (this is called the 'availability' heuristic), or examples that are harder or take longer to come
to mind.
People often have a tendency to rely on the rst piece of information gathered, this heuristic is called
'anchoring and adjustment' - During decision making, anchoring occurs when individuals use an initial piece
of information to make subsequent judgments. People might adjust away from the anchor to get their nal
answer, which would be the logical thing to do; however studies show people tend rely on the rst piece of
information - whether it is right or not (instead of using it as evidence and explain away from it when the
information is false)
So it depends on the cirumstance if people try or don't try to explain (adjust) away from an incorrect
piece of evidence. They might try to justify the rst piece of information oered (the anchor) even though
it wouldn't be the logical thing to do.
So this relates to thinking logically - when weighing evidence, people need to consider if they are being
falsely inuenced by information and are biasing dierent pieces of information in their mind. They might
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
4
be biasing the rst piece of information oered 'the anchor' and be relying too heavily on that instead of
looking more objectively at all of the evidence.
So how exactly does the human mind weigh dierent pieces of information or construct an argument
based o of evidence? It uses mental models to 'model' an argument, I would say. So there are dierent
ways material or evidence can be considered by your mind, and these mental models weigh this evidence
dierently each time. Depending on the set of material or evidence, your mind might consider it dierently
(a 'mental model').
How could someone learn to reason more logically? I just explained two heuristics and how they eect
thinking - by the speed and order of information made available to your mind. People bias the information
they are given or don't consider it logically in many cases, but all that could be done about that to become
a more logical thinker would be to be aware of your personal biases and be more reective.
Hypothetical reasoning
What is hypothetical reasoning? It is creating imaginary worlds to test out our thinking. Here Stanovich
3
explains this type of reasoning in terms of carrying out goals, though I would say this type of thinking is
critical for more complex thought as well:
When we reason hypothetically, we create temporary models of the world and test out actions (or
alternative causes) in that simulated world. In order to reason hypothetically we must, however, have
one critical cognitive capabilitythe ability to distinguish our representations of the real world from
representations of imaginary situations. For example, in considering an alternative goal state dierent
from the one we currently have, we must be able to represent our current goal and the alternative goal
and to keep straight which is which. Likewise, we need to be able to dierentiate the representation of
an action about to be taken from representations of potential alternative actions we are considering.
But the latter must not infect the former while the mental simulation is being carried out.
If you think about it, humans must have a large imaginary world in their minds where they think and test
out what they are thinking.
This probably applies to everything - if you are trying to gure out which
team is going to win a soccer match you might simulate the game in your head. If you are thinking about
anything, you simulate the emotions, actions, behaviors, mathematical equations, or whatever it is - and
this helps you think about it.
Heuristic vs. Rule-based processing
Heuristic processing is low-level, more unconscious and doesn't require as much thought as systematic processing.
4
Systematic processing requires active, careful scrutiny of relevant information and is more cogni-
tively taxing.
Heuristic processing makes use of low-level decision rules such as 'analysts are always right' or 'statistics
don't lie'. However, even though that type of processing makes use of rules, it is a lower-level processing
than when rules are used by the systematic type of processing - which is more cognitive and leads to attitude
change that is more enduring (because it is more conscious).
These dierent ways of processing are related to conscious and unconscious processing, or what is called
in psychology a 'dual process theory' which provides an account of how a phenomenon can occur in two
dierent ways, or as a result of two dierent processes.
Often, the two processes consist of an implicit
(automatic), unconscious process and an explicit (controlled), conscious process.
So rule-based processing usually refers to higher-level logic and casual inference. It follows rules, instead
of merely conforming to them like how weight conforms to the law of gravity. So the unconscious could be
considered to be doing its own thing, however the conscious mind actively thinks and therefore 'consciously'
follows rules or thinks more about rules, more so than simply using a rule as a guideline. An example would
be the rule-based decision rule example I used before to explain heuristic processing. If the rule or thought
3 Stanovich, K. E., + Stanovich, P. J. (2010). A framework for critical thinking, rational thinking, and intelligence. In
D. Preiss + R. J. Sternberg (Eds.), Innovations in educational psychology: Perspectives on learning, teaching and human
development (pp. 195-237). New York: Springer.
4 Chaiken, S. (1980). Heuristic Versus Systematic Information Processing and the Use of Source Versus Message Cues in
Persuasion. Journal of Personality + Social Psychology, 39(5), 752-766. Retrieved from SocINDEX database.
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
5
is 'analysts are always right' then your mind might unconsciously follow that when listening to an analyst
and then you would believe that he or she is right. However if the process is more conscious then you might
think 'well maybe this person is wrong'. The rule wouldn't be as unconscious.
Anyone could really dene 'heuristic processing' as being conscious or unconscious, controlled or automatic actually.
Dierent people have termed the processes of the conscious mind and the processes of
the unconscious mind dierently - these are called 'dual process' theories.
Here Moshman
5
lists all the
combinations of the dierent types of processing as possibilities:
Central to S+W's analysis is a distinction between automatic heuristic processing (characteristic of
what they call System 1) and explicit rule-based processing (characteristic of what they call System 2).
I believe this dichotomy confounds two orthogonal distinctions. Specically, the distinction between
automatic and explicit processing is conceptually orthogonal to the distinction between heuristic and
rule-based processing. Crossing automatic versus explicit with heuristic versus rule-based suggests four
possible types of processing: (a) automatic heuristic processing (System 1), (b) automatic rule-based
processing (not represented in the Stanovich/West analysis), (c) explicit heuristic processing (also not
represented), and (d) explicit rule-based processing (System 2).
The two types not represented probably weren't because they don't make complete sense - rule-based processing is more conscious and controlled, so saying it is automatic would be putting it in the unconscious
category - which is possible, however that is not how it is dened. Explicit heuristic processing doesn't necessarily make much sense either because heuristic processing is dened as being automatic and not cognitively
taxing, however explicit or controlled processes are cognitively taxing because they are more deliberate and
conscious.
Conscious vs. unconscious intuitions
In the 'authors response' section of a Stanovich and West article (the same article as the previous quote (the
6
Moshman commentary in that article)
the authors discuss the dierence between intuitive feelings and ideas
and conscious analytic analysis of people. In the article 'System 1' is more unconscious, forms intuitions,
and the conscious mind then acquires these intuitions. They give the example of a statistics instructor who,
though initially draws conclusions about students and infers probability about their personalities ('for whom
the basic probability axioms are not transparent'), he or she eventually becomes no longer able to emphasize
with them. Basically the unconscious, intuitive mind helps form our conscious understanding of people and
of the probability judgments we make:
We agree with Kahneman that some people may make more nuanced System 1 judgments than others,
and that individual dierences in this capability are of some importance. This is related to Teigen's
point that when System 2 analytic abilities fail, well-framed intuitions may come to our assistance in
narrowing the normative/descriptive gap, and the better those intuitions are the narrower the gap. But,
following Reber (1992a; 1992b; 1993), we would conjecture that the variance in these System 1 abilities
might well be considerably lower than the more recently evolved structures of System 2. Note, however,
that this variability could become larger through the mechanism discussed above instantiating of
automatic System 1 algorithms through practice strategically initiated by System 2. Thus, some of the
well framed intuitions referred to by Teigen may well be acquired intuitions having their origins
in capacity- intensive serial processing, yet now having the encapsulated, automatic characteristics of
modular processes. Some statistics instructors, for example, become unable to empathize with their
students for whom the basic probability axioms are not transparent.
The instructor can no longer
remember when these axioms were not primary intuitions.
5 Diversity in reasoning and rationality: Metacognitive and developmental considerations. David Moshman. Commentary in
Stanovich, K. E., + West, R. F. (2000). Individual dierences in reasoning: Implications for the rationality debate? Behavioral
and Brain Sciences, 23, 645-665.
6 Stanovich, K. E., + West, R. F. (2000). Individual dierences in reasoning: Implications for the rationality debate?
Behavioral and Brain Sciences, 23, 645-665.
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
6
It is obvious that the unconscious mind helps forms our conscious understanding. People have two ways of
thinking about the world, one is unconscious and one is conscious. These two systems must interact all of
the time and inuence each other in dierent ways.
Ways of thinking
The algorithmic level of analysis of mind is the level that just analyzes the details of what is occurring - it
doesn't reect and ask 'why' questions. There are dierent types of thinking dispositions, or ways people
think - these ways of analyzing how someone thinks can help determine if a person is thinking rationally or
irrationally. Here is Stanovich + Stanovich (2010):
The dierence between the algorithmic mind and the reective mind is captured in another wellestablished distinction in the measurement of individual dierencesthe distinction between cognitive
ability and thinking dispositions. The former are, as just mentioned, measures of the eciency of the
algorithmic mind. The latter travel under a variety of names in psychologythinking dispositions or
cognitive styles being the two most popular. Many thinking dispositions concern beliefs, belief structure
and, importantly, attitudes toward forming and changing beliefs. Other thinking dispositions that have
been identied concern a person's goals and goal hierarchy. Examples of some thinking dispositions
that have been investigated by psychologists are: actively open-minded thinking, need for cognition
(the tendency to think a lot), consideration of future consequences, need for closure, superstitious
thinking, and dogmatism (Cacioppo, Petty, + Feinstein 1996; Kruglanski + Webster, 1996; Norris +
Ennis, 1989; Schommer-Aikins, 2004; Stanovich, 1999, 2009; Sternberg, 2003; Sternberg + Grigorenko,
1997; Strathman, Gleicher, Boninger, + Scott Edwards, 1994).
The literature on these types of thinking dispositions is vast and our purpose is not to review that
literature here. It is only necessary to note that the types of cognitive propensities that these thinking
disposition measures reect are the tendency to collect information before making up one's mind, to
seek various points of view before coming to a conclusion, to think extensively about a problem before
responding, to calibrate the degree of strength of one's opinion to the degree of evidence available,
to think about future consequences before taking action, to explicitly weigh pluses and minuses of
situations before making a decision, and to seek nuance and avoid absolutism.
In short, individual
dierences in thinking dispositions include assessing variation in people's goal management, epistemic
values, and epistemic self-regulationdierences in the operation of reective mind.
They are all
psychological characteristics that underpin rational thought and action.
So there are bunch of subjective things a human's mind does that determine how it thinks. I mean in any
single situation how could someone think about their entire 'goal hierarchy' or their 'belief structure' ? Does
that matter if the person is open-minded? How much do you need to think about the future consequences
of your actions or weigh the pluses and minuses of a situation? All of these processes are very subjective
and hard to measure on standard IQ tests; however they are all 'psychological characteristics that underpin
rational thought and action'.
4 The Nature of Reasoning
Deductive reasoning is the same as top-down reasoning, where someone looks at generalizations rst and
then gures out what the details of those would be.
Often it could just be a guess what the details are since you might be inferring the details instead of
nding evidence.
I would say that this type of logic can apply to any type of thinking. For instance, even if I am just
moving the mouse of a computer I could think of it in two ways (deductive or inductive). The deductive way
of thinking would be something like 'I am moving the mouse of the computer, my arm causes it to move and
those are the movements it makes' and the inductive way would be 'these are the movements my computer
mouse is making, I must be deciding to move the mouse - I am directing its movements'. With the deductive
method, the idea of you came before the realization you were making detailed movements, and with the
inductive method the opposite occurred.
http://cnx.org/content/m50415/1.17/
OpenStax-CNX module: m50415
7
My example is dierent from more obvious or straightforward examples of deductive and inductive reasoning, where it is clear what the generalization is and what the details are. I would say that it is still a
good example, however, it is just more subjective. The idea 'I am moving the mouse' is the main idea, and
the detail is 'the mouse is moving, those are the movements it is making'. The reason the movements of the
mouse are the details is because that is where more description can be described.
That example of deductive reasoning is subjective, however. A more typical example would be one with
a concretely broad idea and it would be clear that detail was inferred from it. With my example, however,
you could say that the opposite is true and that the general idea is that 'the mouse is moving', and the
detail is that 'I am moving the mouse' (instead of the opposite). It looks like it depends on which idea comes
rst. Whichever idea comes rst the human mind would assess is the more generalized idea that needs to be
supported. - That idea itself is signicant because people could form delusions, or imagine what the details
are or what the other side of the story is simply because they heard one idea rst.
http://cnx.org/content/m50415/1.17/